Gone are the days when Google would return a simple list of 10 blue links, each neatly packaged with a URL, blue link, and text snippet. Today, search-engine-results pages (SERPs) are far more complex. The majority of SERPs on major web-search engines like Google and Bing present at least one informational, interactive feature.

Left: A results page for George Brett from 2009; Right: A results page for the same query in 2019, with a variety of content types, features, and presentations.

The continuously evolving layout of the SERP is shaping how people search. Each new feature affects the distribution of users' attention on the page. In the old days of web search, users would reliably focus their attention on the first few results at the top of the page and would sequentially move from result to result down the list. (In the research for our first edition of the How People Read on the Web report, we found that in 59% of cases, people scanned the SERP sequentially, from the first results, to the second, and so on, without skipping any results or looking at the right side of the page.)

That linear SERP pattern still exists today, but it’s the exception rather than the rule. Today, we find that people’s attention is distributed on the page and that they process results more nonlinearly than before. We observed so much bouncing between various elements across the page that we can safely define a new SERP-processing gaze pattern — the pinball pattern.

In a pinball pattern, the user scans a results page in a highly nonlinear path, bouncing around between results and SERP features.

A traditional pinball machine arcade game (left) is a glass-covered cabinet using an angled play field with various bumpers, obstacles, and targets. The player uses flippers to shoot and bounce a metal ball around the play field, receiving points for hitting different targets. In a pinball pattern (right), the user’s gaze similarly “bounces” around between visual elements and keywords in a SERP.

Our findings come from the Search Meta-Analysis Project, an analysis of 471 queries made by participants in usability-testing and eyetracking studies that we conducted between 2017 and 2019.

What Causes the Pinball Pattern

Today’s SERPs often involve not only links, but also images, video, embedded text content, and even interactive features. Any given search can return an assortment of different visual elements. The variety of information and presentation plays a critical role in shifting user attention across the SERP.

One participant was researching how to install French drains in her backyard by herself. Her search for french drain returned text results, a shopping-results carousel, a knowledge panel, a people-also-ask element, and a video pack. Her gaze followed a pinball pattern — bouncing around as salient visual elements and keywords redirected her attention.

The visual weight of elements on the page drives people’s scanning patterns. Because these elements are distributed all over the page and because some SERPs have more such elements than others, people’s gaze patterns are not linear. The presence and position of visually compelling elements often affect the visibility of the organic results near them.

This gazeplot shows a study participant’s eye gazes as he searched for the best refrigerator to buy. The numbers represent the order of the fixations. The participant’s gaze flicked around between highly salient elements. First, he focused on the prices in the sponsored shopping results on the right, and then skipped down and to the left to read the featured snippet. He then bounced back up to result #3 and #1 (both ads) before glancing down to the People Also Ask element and to result #6 (the first organic result).

The complex, dynamic content on result pages gets a lot of attention. When SERP features (like featured snippets) were present on a SERP in our study, we found that they received looks in 74% of cases (with a 95% confidence interval of 66–81%).

Even organic results are more complex than they used to be. Some results are presented with sublinks to different sections of the site, as well as a site-specific search box.

A search for restoration hardware brings up the official site as the first result. The result includes six sublinks to popular sections of the site, as well as a site-specific search box.
Searching for dresser in the site-specific search box brings up a new Google SERP with the search-engine command, site:restorationhardware.com ;added to the query. Displaying the command in the search box is a good method to raise awareness of advanced search commands, which are virtually never used by the general web audience, according to our research. In the 471 queries generated by our participants and studied for this project, not one of them involved a search-engine command or operator.
A study participant searched for john james audubon to learn more about the artist before attending an exhibition of his works. His gazeplot shows a pinball pattern directed by compelling text and keywords more than photos or videos.

Because search-results pages are now so inconsistent from query to query, users are often forced to assess the page before digging in and making a selection. That means that layout of a SERP can determine which links get visibility and clicks.

The inconsistency in SERP layout means that users have work more to process it than they used to. It may be that search engines try to encourage people to explore more than just the first result. Yet, people are fairly fast in choosing a search result— we found that users spent an average of 5.7 seconds considering results before they made their first selection (with a 95% confidence interval from 4.9 to 6.5 seconds).

However, if you’re designing SERPs for your in-site search, it’s a better bet to use a consistent layout that doesn’t change radically from query to query. Your goal should be to empower users to take advantage of your site as a whole, not to capture them within the search section of the site.

You Don’t Always Have to Be First

What does this new gaze pattern mean for digital-product teams and content creators?

In the old days, the advice for search-engine optimization (SEO) used to be “If you’re not first, you’re last.” That’s because at that time you weren’t likely to get many people clicking on your link — or even glancing at it — unless you were at the top of the list. Times have changed; we found a wider distribution of both clicks and looks further down the results pages.

Back in 2006, the first result on any given results page received 51% of clicks. In our project, in contrast, we found that the first position on a SERP (defined as the first item listed under the search box) received only 28% of clicks — almost half, which is a dramatic change in user behavior within the short time of only about a decade. 59% of clicks were concentrated in the first three positions, but lower positions received slightly more clicks than in 2006.

This chart compares the distribution of clicks from our Search Meta-Analysis Project against a similar study conducted in 2006. Remember that, unlike in 2006, the 2016–2018 positions aren’t always just filled by simple text links. They could be SERP features or ads as well.

The type of task played a role in the distribution of clicks. For simple, fact-finding tasks (e.g., What’s the tallest building in the world?), people relied heavily on the first few results on the page and were less likely to consider results further down. For complex, research-based tasks (e.g., How much pea gravel do I need to cover my backyard?), people were willing to dig through and consider lower results.

This stacked bar chart compares the proportion of clicks above and below the fold for research tasks and simple tasks (fact-finding and navigation). Only 5% of selections in fact-finding and navigation tasks were below the fold, while 20% of selections in research tasks were below the fold. Both scenarios confirm the disproportionately high importance of information above the fold on web pages. This difference is statistically significant.

This result means that if in-depth, learning-about-a-topic type of tasks bring people to your site, you may be ok if you’re a little further down the SERP.

In our eyetracking research in 2010, we found that in 59% of cases, people didn’t look beyond the 3rd result on the page. In our more recent project, we found that even the 6th position received looks in 36% of cases.

This chart shows the proportion of cases in which each SERP position received a look or a click. For example, the first position received 19% of all looks and 28% of all clicks.

These results are encouraging for site teams. If your site isn’t ranking in the first position, that isn’t ideal. But as long as you’re appearing near the top (within the first 5 results), you may have around a 10–20% chance of getting a click and anywhere from a 40–80% chance of getting a look.

A glance may sound unimportant, but it has the potential to be quite valuable. In our research, we often observe users who discover previously unknown sites during search. Even if users don’t click on a link to your site the first time they see it, simply seeing your site name provides awareness and increases familiarity, thus bolstering your chances for next time.

We observed several instances where users clicked a result and, while that page was loading, continued to scan the SERP (a phenomenon we call perpetual viewing). If their selected result didn’t answer their question, some users returned to the SERP and chose another result they’d been eyeing earlier.

One piece of classic SEO advice does remain true — you need to appear on the first page of results. In our analysis, users only clicked past the first page of results in 2% of queries.

Additionally, the complexity of SERP elements means more competition for content creators. For simple information-seeking tasks, users often answer their questions on the SERP itself, without clicking on any result at all.

Conclusion

  • When search results pages contain complex and visually attractive elements, users are more likely to be drawn to those elements and distribute their attention across the SERP.
  • If you can make it into the top 5 positions on a SERP, you’ll have a good chance (40–80%) of getting a valuable glance from your user.
  • It’s still important to appear within the first page of results, since people still aren’t likely to click through to the second page.
  • Consider adding some of these nontraditional SERP features to your in-site search, if it makes sense for your content. But remember that when results pages are very inconsistent from query to query, users have to re-assess the page each time, which adds to interaction cost.

References

Hoa Loranger and Jakob Nielsen, Prioritizing Web Usability (2006).

Kara Pernice, Kathryn Whitenton, and Jakob Nielsen, How People Read on the Web: The Eyetracking Evidence (2013).

Jakob Nielsen and Kara Pernice, Eyetracking Web Usability (2010).